Classification of older age patients with Type 2 Diabetes Mellitus in terms of severity and glucose lowering therapy needs using Latent Class Analysis.

Christiaens, Antoine [UCL] Hermans, Michel [UCL] Boland, Benoît [UCL] Henrard, Séverine [UCL]

INTRODUCTION. Older patients with Type 2 Diabetes Mellitus (T2DM) represent a heterogeneous group regarding their metabolic profile. It complicates glucose-lowering-therapy’s (GLT) choice and management, as it should be adjusted to this severity of T2DM. This study aimed to identify subgroups existing among older patients with T2DM. METHODS. Retrospective study of an observational cohort of outpatients followed in a Belgian diabetic clinic. Included patients (n=147; age=80.16±3.97; female=37%) were all diagnosed with T2DM, age ≥75years and benefited of a Homeostasis Model Assessment (HOMA). A latent class analysis was conducted using HOMA variables (Insulin sensitivity (S), β-cell function (β) and hyperbolic product (βxS)) and age at diagnosis to classify patients. RESULTS. The resulting model classified patients into 6 subgroups. Patients in groups 1 (10.9%) and 6 (9.5%) had the oldest age at diagnosis with a preserved βxS. From Groups 3 to 5, insulin sensitivity decreased, but β-cell function increased inversely, resulting in a moderate decrease in βxS in all 3 groups. Group 2 (25.2%) had an impaired β-cell function and the lowest βxS. Mean total Defined Daily Dose of GLT was <1.0 in Groups 1 and 6, between 1.2 and 1.6 in Groups 3 to 5 and of 2.4 in Group 2. In the same way, other metabolic differences were found between groups. CONCLUSION. This study identified 6 clinically subgroups in terms of severity and metabolic profile among T2DM older patients. Intensity and choice of GLT should be adapted on this basis, in addition to the other existing criteria of treatment individualization.